Identificação e Elaboração de Variáveis Estatísticas Através de Técnicas Multivariadas no Estudo da Produção de Grãos e Fibras, na Região Administrativa de Presidente Prudente-SP

Abstract

This work deals with the study to obtain statistical variables through multivariate techniques applied to the data of grain and fiber production in the Administrative Region of Presidente Prudente-SP, and is based on the fact that the region be benefited by having a distribution logistics system three-ways (rail, road and waterway). The used database was extracted from the Survey of Agricultural Production Units (LUPA) and refers to the areas provided by the municipalities studied for the production of grains and fiber and also economic variables related to agribusiness. The applied statistical techniques were factor analysis and cluster analysis. Through factorial analysis were obtained, from 13 original variables, three statistical descriptors (factors). The clusters were made from the scores factorials, finding three groups per descriptor and also by the conjoint of the descriptors. At the end of the work, was characterized the 10th RA State of São Paulo according to the variables selected all of the year 2008, giving basement to auxiliary in elaboration business strategies and formulation of the public policies for the development of the region under study.